1.1 Field of the Invention
The present invention provides new proteins for combatting insects, and particularly, coleopteran, dipteran, and lepidopteran insects sensitive to the disclosed δ-endotoxins derived from Bacillus thuringiensis. The invention provides novel chimeric crystal proteins and the chimeric cry gene segments which encode them, as well as methods for making and using these DNA segments, methods of producing the encoded proteins, methods for making synthetically-modified chimeric crystal proteins, and methods of making and using the synthetic crystal proteins.
1.2 Description of Related Art
1.2.1 B. Thuringiensis CRYSTAL PROTEINS 
The Gram-positive soil bacterium B. thuringiensis is well known for its production of proteinaceous parasporal crystals, or δ-endotoxins, that are toxic to a variety of lepidopteran, coleopteran, and dipteran larvae. B. thuringiensis produces crystal proteins during sporulation which are specifically toxic to certain species of insects. Many different strains of B. thuringiensis have been shown to produce insecticidal crystal proteins, and compositions comprising B. thuringiensis strains which produce proteins having insecticidal activity have been used commercially as environmentally-acceptable insecticides because of their toxicity to the specific target insect, and non-toxicity to plants and other non-targeted organisms.
Commercial formulations of naturally occurring B. thuringiensis isolates have long been used for the biological control of agricultural insect pests. In commercial production, the spores and crystals obtained from the fermentation process are concentrated and formulated for foliar application according to conventional agricultural practices.
1.2.2 NOMENCLATURE OF CRYSTAL PROTEINS 
A review by Höfte et al., (1989) describes the general state of the art with respect to the majority of insecticidal B. thuringiensis strains that have been identified which are active against insects of the Order Lepidoptera, i.e., caterpillar insects. This treatise also describes B. thuringiensis strains having insecticidal activity against insects of the Orders Diptera (i.e. flies and mosquitoes) and Coleoptera (i.e. beetles). A number of genes encoding crystal proteins have been cloned from several strains of B. thuringiensis. Höfte et al. (1989) discusses the genes and proteins that were identified in B. thuringiensis prior to 1990, and sets forth the nomenclature and classification scheme which has traditionally been applied to B. thuringiensis genes and proteins. cry1 genes encode lepidopteran-toxic Cry1 proteins. cry2 genes encode Cry2 proteins that are toxic to both lepidopterans and dipterans. cry3 genes encode coleopteran-toxic Cry3 proteins, while cry4 genes encode dipteran-toxic Cry4 proteins, etc.
Recently a new nomenclature has been proposed which systematically classifies the Cry proteins based upon amino acid sequence homology rather than upon insect target specificities. This classification scheme is summarized in Table 1.
TABLE IREVISED B. THURINGIENSIS δ-ENDOTOXINNOMENCLATUREANewOldGenBank Accession #Cry1AaCryIA(a)M11250Cry1AbCryIA(b)M13898Cry1AcCryIA(c)M11068Cry1AdCryIA(d)M73250Cry1AeCryIA(e)M65252Cry1BaCryIBX06711Cry1BbET5L32020Cry1BcPEG5Z46442Cry1BdCryE1U70726Cry1CaCryICX07518Cry1CbCryIC(b)M97880Cry1DaCryIDX54160Cry1DbPrtBZ22511Cry1EaCryIEX53985Cry1EbCryIE(b)M73253Cry1FaCryIFM63897Cry1FbPrtDZ22512Cry1GaPrtAZ22510Cry1GbCryH2U70725Cry1HaPrtCZ22513Cry1HbU35780Cry1IaCryVX62821Cry1IbCryVU07642Cry1JaET4L32019Cry1JbET1U31527Cry1KU28801Cry2AaCryIIAM31738Cry2AbCryIIBM23724Cry2AcCryIICX57252Cry3ACryIIIAM22472Cry3BaCryIIIBX17123Cry3BbCryIIIB2M89794Cry3CCryIIIDX59797Cry4ACryIVAY00423Cry4BCryIVBX07423Cry5AaCryVA(a)L07025Cry5AbCryVA(b)L07026Cry5BU19725Cry6ACryVIAL07022Cry6BCryVIBL07024Cry7AaCryIIICM64478Cry7AbCryIIICbU04367Cry8ACryIIIEU04364Cry8BCryIIIGU04365Cry8CCryIIIFU04366Cry9ACryIGX58120Cry9BCryIXX75019Cry9CCryIHZ37527Cry10ACryIVCM12662Cry11ACryIVDM31737Cry11BJeg80X86902Cry12ACryVBL07027Cry13ACryVCL07023Cry14ACryVDU13955Cry15A34 kDaM76442Cry16Acbm71X94146Cry17Acbm71X99478Cry18ACryBP1X99049Cry19AJeg65Y08920Cyt1AaCytAX03182Cyt1AbCytMX98793Cyt1BU37196Cyt2ACytBZ14147Cyt2BCytBU52043aAdapted from: http://epunix.biols.susx.ac.uk/Home/Neil_Crickmore/Bt/index.html 1.23 MODE OF CRYSTAL PROTEIN TOXICITY 
All δ-endotoxin crystals are toxic to insect larvae by ingestion. Solubilization of the crystal in the midgut of the insect releases the protoxin form of the δ-endotoxin which, in most instances, is subsequently processed to an active toxin by midgut protease. The activated toxins recognize and bind to the brush-border of the insect midgut epithelium through receptor proteins. Several putative crystal protein receptors have been isolated from certain insect larvae (Knight et al., 1995; Gill et al., 1995; Masson et al., 1995). The binding of active toxins is followed by intercalation and aggregation of toxin molecules to form pores within the midgut epithelium. This process leads to osmotic imbalance, swelling, lysis of the cells lining the midgut epithelium, and eventual larvae mortality.
1.2.4 MOLECULAR BIOLOGY OF δ-ENDOTOXINS 
With the advent of molecular genetic techniques, various δ-endotoxin genes have been isolated and their DNA sequences determined. These genes have been used to construct certain genetically engineered B. thuringiensis products that have been approved for commercial use. Recent developments have seen new δ-endotoxin delivery systems developed, including plants that contain and express genetically engineered δ-endotoxin genes.
The cloning and sequencing of a number of δ-endotoxin genes from a variety of Bacillus thuringiensis strains have been described and are summarized by Höfte and Whiteley, 1989. Plasmid shuttle vectors designed for the cloning and expression of δ-endotoxin genes in E. coli or B. thuringiensis are described by Gawron-Burke and Baum (1991). U.S. Pat. No. 5,441,884 discloses a site-specific recombination system for constructing recombinant B. thuringiensis strains containing δ-endotoxin genes that are free of DNA not native to B. thuringiensis. 
The Cry1 family of crystal proteins, which are primarily active against lepidopteran pests, are the best studied class of δ-endotoxins. The pro-toxin form of Cry1 δ-endotoxins consist of two approximately equal sized segments. The carboxyl-half, or pro-toxin segment, is not toxic and is thought to be important for crystal formation (Arvidson et al., 1989). The amino-half of the protoxin comprises the active-toxin segment of the Cry1 molecule and may be further divided into three structural domains as determined by the recently described crystallographic structure for the active toxin segment of the Cry1Aa δ-endotoxin (Grochuiski et al., 1995). Domain 1 occupies the first third of the active toxin and is essential for channel formation (Thompson et al., 1995). Domain 2 and domain 3 occupy the middle and last third of the active toxin, respectively. Both domains 2 and 3 have been implicated in receptor binding and insect specificity, depending on the insect and δ-endotoxin being examined (Thompson et al., 1995).
1.2.5 CHIMERIC CRYSTAL PROTEINS 
In recent years, researchers have focused effort on the construction of hybrid δ-endotoxins with the hope of producing proteins with enhanced activity or improved properties. Advances in the art of molecular genetics over the past decade have facilitated a logical and orderly approach to engineering proteins with improved properties. Site-specific and random mutagenesis methods, the advent of polymerase chain reaction (PCR™) methodologies, and the development of recombinant methods for generating gene fusions and constructing chimeric proteins have facilitated an assortment of methods for changing amino acid sequences of proteins, fusing portions of two or more proteins together in a single recombinant protein, and altering genetic sequences that encode proteins of commercial interest.
Unfortunately, for crystal proteins, these techniques have only been exploited in limited fashion. The likelihood of arbitrarily creating a chimeric protein with enhanced properties from portions of the numerous native proteins which have been identified is remote given the complex nature of protein structure, folding, oligomerization, activation, and correct processing of the chimeric protoxin to an active moiety. Only by careful selection of specific target regions within each protein, and subsequent protein engineering can toxins be synthesized which have improved insecticidal activity.
Some success in the area, however, has been reported in the literature. For example, the construction of a few hybrid δ-endotoxins is reported in the following related art: Intl. Pat. Appl. Publ. No. WO 95/30753 discloses the construction of hybrid B. thuringiensis δ-endotoxins for production in Pseudomonas fluorescens in which the non-toxic protoxin fragment of Cry1F has been replaced by the non-toxic protoxin fragment from the Cry1Ac/Cry1Ab that is disclosed in U.S. Pat. No. 5,128,130.
U.S. Pat. 5,128,130 discloses the construction of hybrid B. thuringiensis δ-endotoxins for production in P. fluorescens in which a portion of the non-toxic protoxin segment of Cry1Ac is replaced with the corresponding non-toxic protoxin fragment of Cry1Ab. U.S. Pat. No. 5,055,294 discloses the construction of a specific hybrid δ-endotoxin between Cry1Ac (amino acid residues 1-466) and Cry1Ab (amino acid residues 466-1155) for production in P. fluorescens. Although the aforementioned patent discloses the construction of a hybrid toxin within the active toxin segment, no specifics are presented in regard to the hybrid toxin's insecticidal activity. Intl. Pat. Appl. Publ. No. WO 95/30752 discloses the construction of hybrid B. thuringiensis δ-endotoxins for production in P. fluorescens in which the non-toxic protoxin segment of Cry1C is replaced by the non-toxic protoxin segment from Cry1Ab. The aforementioned application further discloses that the activity against Spodoptera exigua for the hybrid δ-endotoxin is improved over that of the parent active toxin, Cry1C.
Intl. Pat. Appl. Publ. No. WO 95/06730 discloses the construction of a hybrid B. thuringiensis δ-endotoxin consisting of domains 1 and 2 of Cry1E coupled to domain 3 and the non-toxic protoxin segment of Cry1C. Insect bioassays performed against Manduca sexta (sensitive to Cry1C and Cry1E), Spodoptera exigua (sensitive to Cry1C), and Mamestra brassicae (sensitive to Cry1C) show that the hybrid Cry1E/Cry1C hybrid toxin is active against M. sexta, S. exigua, and M. brassicae. The bioassay results were expressed as EC50 values (toxin concentration giving a 50% growth reduction) rather than LC50 values (toxin concentration giving 50% mortality). Although the δ-endotoxins used for bioassay were produced in B. thuringiensis, only artificially-generated active segments of the δ-endotoxins were used, not the naturally-produced crystals typically produced by B. thuringiensis that are present in commercial B. thuringiensis formulations. Bioassay results indicated that the LC50 values for the hybrid Cry1E/Cry1C crystal against S. frugiperda were 1.5 to 1.7 fold lower (more active) than for native Cry1C. This art also discloses the construction of a hybrid B. thuringiensis δ-endotoxin between Cry1Ab (domains 1 and 2) and Cry1C (domain 3 and the non-toxic protoxin segment), although no data are given regarding the hybrid toxin's activity or usefulness.
Lee et al. (1995) report the construction of hybrid B. thuringiensis δ-endotoxins between Cry1Ac and Cry1Aa within the active toxin segment. Artificially generated active segments of the hybrid toxins were used to examine protein interactions in susceptible insect brush border membranes vesicles (BBMV). The bioactivity of the hybrid toxins was not reported.
Honee et al. (1991) report the construction of hybrid δ-endotoxins between Cry1C (domain 1) and Cry1Ab (domains 2 and 3) and the reciprocal hybrid between Cry1Ab (domain 1) and Cry1C (domains 2 and 3). These hybrids failed to show any significant increase in activity against susceptible insects. Furthermore, the Cry1C (domain 1)/Cry1Ab (domains 2 and 3) hybrid toxin was found to be hypersensitive to protease degradation. A report by Schnepf et al. (1990) discloses the construction of Cry1Ac hybrid toxin in which a small portion of domain 2 was replaced by the corresponding region of Cry1Aa, although no significant increase in activity against susceptible insect larvae was observed.
1.3 DEFICIENCIES IN THE PRIOR ART 
The limited successes in producing chimeric crystal proteins which have improved activity have negatively impacted the field by thwarting efforts to produce recombinantly-engineered crystal protein for commercial development, and to extend the toxic properties and host specificities of the known endotoxins. Therefore, what is lacking in the prior art are reliable methods and compositions comprising recombinantly-engineered crystal proteins which have improved insecticidal activity, broad-host-range specificities, and which are suitable for commercial production-in B. thuringiensis. 